Figure 3 depicts the combined incidence of hospitalization for CONUS vs Europe by calendar year. There is consistent parallel movement in the absolute level of hospitalizations for the two sites, with a relatively constant 30 to 40 percent higher incidence in the latter. The remaining figures demonstrate that there is considerable heterogeneity among sex, race, and age groups. Figure 4 for white male subjects and a similar one for white female subjects (not shown) reveal that the total rate observed previously is reflected in those white subjects older than 24 yr; but in both sexes, rates for those less than this age show a consistent gradual decrease over the study period; and, furthermore, note the similarity of rates between CONUS and Europe in the younger category. Figure 5 for black male subjects and a similar one for black female subjects (not shown) reveal that black subjects are worse off in Europe for all ages, with no discernible trend in those less than 24 yr but with a very noticeable increasing trend in both sexes for those in the older group. generic doxycycline
During this same interval, there were ten deaths attributable to asthma. Six of these occurred in Europe, whereas four occurred in CONUS. Of these, two deaths in CONUS and one in Europe occurred in individuals greater than 34 yr of age and are not included in subsequent calculations. The respective denominators are 3,852,913 and 8,397,890 person-yr, giving incidence densities of 0.13 and 0.024 per 100,000 person-yr of observation. The exact binomial probability of observing five or more deaths, out of seven total, in Europe, based on the fraction of total population time, is 0.035. Applying the rates in CONUS to the population time in Europe indicates that about three of those deaths may be attributable to exposure to the European environment, or an etiologic fraction of 77 percent. Similarly, there were 1,277 total hospital discharges in Europe, compared to 1,938 in CONUS. The incidence densities are 33.14 and 23.08 per 100,000 person-yr, respectively; and using the normal approximation to the binomial, the difference is highly significant (p<0.001). The etiologic fraction is just under half of that for the previously mentioned mortality at 30.4 percent and corresponds quite well to the observed difference in incidence of hospitalization depicted in Figure 3. Therefore, approximately 388 hospital episodes in Europe can be attributed to European exposure.
Figure 3. Expected total hospital discharge rates by calendar year for Europe and CONUS.
Figure 4. Expected hospital discharge rates for Europe and CONUS for white male subjects.
Figure 5. Expected hospital discharge rates for Europe and CONUS for black male subjects.